| title | Example Code & Notebooks |
|---|---|
| description | Hands-on tutorials, walkthroughs, and colab notebooks for every W&B product. |
Browse end-to-end examples grouped by product. Each page below links to a tutorial, walkthrough, or page containing a runnable colab notebook.
Cross-product entry point for getting up and running with Weights & Biases. Install W&B and start tracking experiments in minutes. Track experiments, log metrics, and visualize results. Add tracing to your LLM application to debug and monitor model interactions. Trace model calls, compare outputs, and run evaluations using Serverless Inference. Build a leaderboard to compare models and experiments. Intro to tracking metrics, hyperparameters, system metrics, and model artifacts. Save experiment configuration with a dictionary-like object. Explore run data in interactive workspaces. Log rich media: 3D point clouds, molecules, HTML, histograms, and more. Log model artifacts to a run with `run.log_model()` and `run.use_model()`. Build and track plots from ML experiments. Intro to hyperparameter search and model optimization with Sweeps. Define, initialize, and run a hyperparameter sweep. Intro to W&B Artifacts and how to get started. Create, track, and use a dataset artifact across experiments. Set TTL policies on artifacts to manage storage. Iterate on datasets and understand model predictions with Tables. Log tables, visualize, and query structured data. Compare, filter, group, and sort tables in merged or side-by-side views. Manage and share artifact versions across your organization. Use default, custom, and protected aliases in the Registry. Project management and collaboration tools for ML projects. Create a W&B Report with the App UI or programmatically. Edit reports interactively or with the Report API. Build custom charts in W&B projects with Vega visualizations. Use custom charts to build tailored visualizations in the W&B UI. Use the embedding projector to explore object embeddings. Track experiments, checkpoint models, and visualize predictions with Keras callbacks. Track metrics, gradients, and models with the PyTorch integration. Use the built-in `WandbLogger` with PyTorch Lightning. Automatically log training metrics, model parameters, and configs. Track LLM fine-tuning experiments with the torchtune WandBLogger. Log custom metrics, use estimator hooks, and sync TensorBoard logs. Log gradient boosting metrics, feature importance, and model performance. Use the built-in W&B integration in YOLOv5 for experiment tracking and versioning. Visualize and track Hugging Face model performance with W&B. Use W&B with the Hugging Face Transformers Trainer. Integrate W&B with Hugging Face Simple Transformers. Autolog prompts, generated media, and pipeline architecture. Log chat completions, fine-tuning jobs, and token usage metrics. Fine-tune Azure OpenAI models with W&B experiment tracking. Build an evaluation pipeline with Weave Models and Evaluations. Build and evaluate a RAG application with LLM judges. Track deeply nested call structures with W&B tracing. Track and version your application and its parameters with Weave Model. A beginner-friendly introduction to tracing with Weave. Get hands-on with running evaluations in Weave. Run evaluations on Hugging Face datasets with Weave. Load a CSV into a Weave dataset and use it in evaluations. Combine W&B Models and Weave in a single workflow. Implement chain-of-density prompting for iterative summarization. Optimize prompts with DSPy and track results in Weave. Route between models dynamically with NotDiamond. Coordinate multiple agents that produce structured output. Build and evaluate a code-generation pipeline with Weave. Trace and evaluate a computer-vision OCR pipeline. Work with audio inputs and outputs in Weave traces. Monitor a production LLM application with Weave. Collect and act on user feedback from production traffic. Use Weave scorers as guardrails for production LLM calls. Track custom per-model costs alongside traces. Redact PII in Weave traces for sensitive workloads. Call the Weave Service API directly to record traces. Use small local language models to evaluate AI safety and quality. Route traces to domain experts and export structured feedback. Track and manage costs for LLM operations. Automatically track and log LLM calls made via the Anthropic SDK. Automatically track and log LLM calls made via the Cohere Python library. Trace and log Google GenAI model calls. Track and monitor Groq LPU inference with Weave. Trace and evaluate Mistral AI model calls with Weave. Integrate OpenAI with Weave for tracing, evaluation, and monitoring. Automatically track and log LLM calls made via LiteLLM. Monitor and trace multi-agent applications with CrewAI. Track and log calls made using DSPy modules and functions. Trace structured-output calls made via Instructor. Track and log all calls made through the LangChain Python library. Use the Verdict evaluation framework to monitor LLM evaluation pipelines. Track and analyze ML applications with Hugging Face Hub. Trace activity between your MCP client and MCP server. Fine-tune and deploy a LoRA adapter with Serverless Inference. Integrate Cline with the Serverless Inference endpoints. Post-train models with reinforcement learning. Fine-tune models with Serverless SFT using the OpenPipe ART framework. Make inference requests to the models you've trained. Train a PyTorch model in a Serverless Sandbox environment. Invoke an OpenAI agent within a Serverless Sandbox.